CN106874108B - Technology for minimizing number of micro clouds in mobile cloud computing - Google Patents

Technology for minimizing number of micro clouds in mobile cloud computing Download PDF

Info

Publication number
CN106874108B
CN106874108B CN201611234023.4A CN201611234023A CN106874108B CN 106874108 B CN106874108 B CN 106874108B CN 201611234023 A CN201611234023 A CN 201611234023A CN 106874108 B CN106874108 B CN 106874108B
Authority
CN
China
Prior art keywords
micro
clouds
cloud
user
access points
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CN201611234023.4A
Other languages
Chinese (zh)
Other versions
CN106874108A (en
Inventor
马隆杰
武继刚
陈龙
刘竹松
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangdong University of Technology
Original Assignee
Guangdong University of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guangdong University of Technology filed Critical Guangdong University of Technology
Priority to CN201611234023.4A priority Critical patent/CN106874108B/en
Publication of CN106874108A publication Critical patent/CN106874108A/en
Application granted granted Critical
Publication of CN106874108B publication Critical patent/CN106874108B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46Multiprogramming arrangements
    • G06F9/50Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/23Clustering techniques
    • G06F18/232Non-hierarchical techniques
    • G06F18/2321Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions

Abstract

In order to improve the user experience of mobile application in a mobile cloud computing environment and save the operation cost of a micro cloud provider, how to use the minimum number of micro clouds to meet the user delay requirement has attracted wide attention of the micro cloud provider. The invention discloses a resource allocation and placement method for meeting the service delay requirement of a user by using the minimum number of micro clouds. According to the method, the number of the micro clouds is increased circularly, all micro cloud placement positions are selected again by adopting the idea of a clustering algorithm in each micro cloud increasing process, user requests are distributed, and finally the number of resources required by the micro clouds and the average delay distributed to the micro clouds by users are calculated according to the user request resources distributed to the micro clouds. After the user converts the network topology and the user request information of the current network access point into corresponding data to be used as the input of the algorithm, the algorithm obtains the minimum needed number of the micro-clouds, the proper placement positions of the micro-clouds and the needed number of resources through corresponding calculation.

Description

Technology for minimizing number of micro clouds in mobile cloud computing
Technical Field
The invention relates to a computer technology, in particular to a resource allocation and placement technology for minimizing the number of micro clouds in mobile cloud computing.
Background
Today, mobile devices have become ubiquitous in people's lives, and mobile applications have become more sophisticated and have increased device resource demands in order to provide mobile device users with more services. However, mobile devices tend to be small in size based on their portability considerations, and thus the battery life, computing power, network bandwidth, and storage resources of the mobile devices tend to be limited. The contradiction between the increasing resource demand of mobile applications and the shortage of mobile device resources becomes more and more prominent, in order to improve the running efficiency of the mobile application on the mobile device, it is proposed to upload the workload of the mobile application to a resource-rich cloud server for execution, however, in the traditional mobile cloud computing architecture, the longer distance between the mobile user and the cloud center causes the user uploading task to generate higher endpoint transmission delay, high transmission delay of delay-sensitive applications (such as augmented reality) is fatal, and in order to reduce communication delay of mobile applications for obtaining services from a cloud, people adopt micro-cloud as a new element to expand a mobile device cloud architecture, the micro-cloud is a computer or a computer cluster which is rich in resources and stable, the mobile applications can upload services to a relatively close micro-cloud for processing, and delay of obtaining the services from a remote cloud center is effectively reduced. Therefore, the placement position of the micro-cloud equipment and the reasonable allocation of resources have important significance for reducing the delay of the mobile application.
In the prior art, resources and quantity of micro clouds can be preset, then a proper position is selected for placement through a corresponding algorithm, and finally a mobile application request is distributed to a closer micro cloud for execution so as to achieve the purpose of minimizing average delay of users. However, in a real-world environment, the distribution density of mobile devices is different, the resources requested by mobile applications are also different, artificial setting of resources of a clout may cause uneven allocation of resources of the clout, so that the resources are not effectively utilized, and the mobile application requests are allocated to a nearest clout for execution to reduce the request delay, and if the nearest clout is insufficient, the mobile application requests may only wait or be allocated to a distant clout for execution, which may cause the delay of the application to increase. Therefore, it is very important to select a proper position to place the micro cloud and reasonably allocate micro cloud resources to effectively utilize resources and reduce user request delay to improve user experience. In addition, for the cloud service provider, it is of great significance to reasonably place the micro-clouds and perform resource allocation under the condition of ensuring the average delay of the user request, so as to reduce the deployment number of the micro-clouds and save the deployment cost.
Disclosure of Invention
The invention aims to overcome the defects and shortcomings of the prior art, and provides an effective micro cloud resource allocation and deployment method aiming at the problem of unreasonable micro cloud resource allocation in a real network environment, so that the micro cloud is allocated and placed under the condition that the average delay of a user request is not more than a given delay value, and the technology for minimizing the number of used micro clouds is achieved.
The main idea of the invention is as follows:
firstly, a network topology of an area where micro clouds are to be deployed is converted into a network topology graph (as shown in fig. 1), nodes which can be placed by the micro clouds in the topology graph are used as user request access points, the access points are connected with the access points by using weighted edges, the weight value of the edges represents communication delay, the delay between the two access points is the length of a shortest path between the two access points, each access point has different numbers of user requests, and resources required by each user request are possibly different. When the same strategy for placing the micro clouds and a user distribution method are adopted in a network with fixed number of access points to place the micro clouds, the average delay distributed to the micro clouds by a user request is reduced along with the increase of the number of the placed micro clouds, therefore, the algorithm of the invention gradually increases the number K of the micro clouds in an iterative mode according to the average delay D input by the user, the position for placing the micro clouds is selected in the access points of the graph by adopting the thought of a K-MEDOIDS, after the micro clouds are placed, the user request connected to each access point is distributed to the micro cloud nearest to the access point to ensure that the average delay of the user obtained by the current number of the micro clouds is minimum, when the average delay obtained by the increase of the number K of the micro clouds is less than or equal to the given delay, the resources required by the micro clouds at each placing position are calculated according to the user request distributed in the micro clouds, k is the required minimum number of the micro clouds.
The purpose of the invention is realized by the following technical scheme
1. The algorithm is implemented in C + + or other programming languages.
2. And designing a reasonable user network data input interface according to the user requirements.
Compared with the prior art, the invention has the following advantages and effects:
the method can calculate the required minimum number of the micro-clouds under the condition of giving the average delay of a user in a large-scale network environment, select reasonable placement for each micro-cloud for placement, and finally calculate the resources required by the micro-clouds at all positions. The micro cloud service provider can reasonably distribute micro cloud resources, meet the user requirements and simultaneously minimize the deployment operation cost.
Drawings
Fig. 1 is a diagram of a network model.
Fig. 2 is a system flow diagram.
Fig. 3 is an algorithm flow chart.
FIG. 4 is algorithm pseudo code.
Detailed Description
The present invention will be described in further detail with reference to examples and drawings, but the present invention is not limited thereto.
Examples are given.
The invention provides an effective micro cloud resource allocation and deployment algorithm based on the problem of micro cloud resource allocation and placement in mobile cloud computing. The present invention can be deployed in the following manner for simplicity of deployment. Firstly, the algorithm can be deployed in a system mode, the system comprises a client and a server, the server can be located locally or in a remote server, the client is responsible for converting an actual network environment into a network extension and uploading the network extension and the uploading to the server, and the server is responsible for calculating the placement position of the micro cloud and the resource quantity required by the micro cloud and returning the calculation result to the client, so that the communication difficulty and the expense of calculation time in the data storage and retrieval process are greatly reduced.
The method comprises the following concrete steps:
firstly, a user converts an actual network environment into a network supplement graph (as shown in fig. 1), wherein edges between every two access points in the graph contain weights which can represent endpoint delay, transmission cost and the like, the user can convert the network supplement graph into an edge weight matrix which is used as algorithm input, the number of user requests on each access point and resources required by each user request can be stored by using a corresponding data type array, and then the user inputs corresponding data into a system through a client.
After the input is finished, the user can upload data to the server through the client to perform calculation, and after the calculation of the server is finished, the placing positions of the micro clouds and the resource number required by the micro clouds at all the positions generate corresponding results and return the results to the user.
The data structure of the network diagram and the parameter values of the access points related in the specific implementation mode of the invention can be reasonably designed according to the actual environment requirements. Reasonable modifications in implementation detail can be made by those skilled in the art without departing from the scope of the invention. The above embodiments are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents thereof, and all such changes, modifications, substitutions, combinations, and simplifications are intended to be included in the scope of the present invention.

Claims (3)

1. A method for determining and placing resources of a micro cloud in mobile cloud computing is characterized by comprising the following specific steps:
s1, giving an average delay D of one user, wherein the number K of the micro clouds is 0;
s2, number of micro clouds K ═ K +1
S3, randomly selecting K access points as micro cloud placement positions;
s4, clustering the access points where the micro clouds are placed;
s5, for each class adjustment class center, namely the placing position of the micro cloud, judging whether the placing positions of all the micro clouds are changed, if not, jumping to S6, otherwise, jumping to S4;
s6, calculating the resources needed by each micro cloud and the average delay avgD distributed to the micro cloud by all users according to the clustering result in S5, jumping to S2 if the avgD is larger than or equal to D, otherwise, outputting the result;
the step S2 further includes randomly selecting K access points as initial placement positions of the cloudiness, where each position can only place one cloudiness, and each cloudiness can only place one position, and the step S2 randomly selects K access points as initial placement positions of the cloudiness; and, in said step S3, all the access points are classified into K classes, and classified into the closest clouding for each access point; each access point is classified into the micro cloud closest to the access point, so that the phenomenon of non-uniform clustering and unreasonable classification in the clustering process is avoided; after dividing all the access points into K classes in step S4, calculating a total delay allocated to the node by all the user requests of the class to which the primary node belongs for each access point of each class, and selecting the node with the minimum total delay as a new micro cloud placement position of the class; step S4 is repeated in step S5 until the positions of the K micro clouds are no longer changed, the final position is the placement position of the micro clouds, and the sum of all the user request resources of the class to which the micro clouds belong is the resource required by the micro clouds; in the step S6, the number of the micro clouds is cyclically increased, and the calculated placement position of the micro clouds and the average delay of the user to the micro clouds are ensured to be minimum under the current number of the micro clouds, and when the number of the placed micro clouds is increased to meet the average delay of the user, the number of the micro clouds that are used at the minimum is obtained.
2. The method of claim 1, wherein: selecting a micro-cloud placement position in an access point of a graph by utilizing the thought of a K-MEDOIDS (K-MedoIDS center point clustering algorithm), after micro-cloud placement is completed, a user request connected to each access point is distributed to a micro-cloud nearest to the access point, then calculating resources required by the micro-cloud placed at each position according to the user request distributed in the micro-cloud, and ensuring that the average delay of the user obtained by the current number of the micro-clouds is minimum, thereby ensuring that the minimum number of the micro-clouds can be calculated by the algorithm under the condition of the given average delay of the user; sampling access points by adopting the idea of Clara algorithm according to the network scale and the actual requirement to replace the whole network, then obtaining the optimal MEDOIDS on the sampled access points by utilizing the K-MEDOIDS algorithm, repeating the processes from S3 to S6 for a plurality of times in the cyclic process of each increase of the micro cloud, and finally classifying and calculating the resources required by the micro cloud for all the access points according to the step 1.
3. Method according to one of claims 1-2, characterized in that the method is applied to resource allocation and/or deployment of base stations, servers.
CN201611234023.4A 2016-12-28 2016-12-28 Technology for minimizing number of micro clouds in mobile cloud computing Expired - Fee Related CN106874108B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201611234023.4A CN106874108B (en) 2016-12-28 2016-12-28 Technology for minimizing number of micro clouds in mobile cloud computing

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201611234023.4A CN106874108B (en) 2016-12-28 2016-12-28 Technology for minimizing number of micro clouds in mobile cloud computing

Publications (2)

Publication Number Publication Date
CN106874108A CN106874108A (en) 2017-06-20
CN106874108B true CN106874108B (en) 2021-02-09

Family

ID=59164066

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201611234023.4A Expired - Fee Related CN106874108B (en) 2016-12-28 2016-12-28 Technology for minimizing number of micro clouds in mobile cloud computing

Country Status (1)

Country Link
CN (1) CN106874108B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109639833B (en) * 2019-01-25 2021-09-07 福建师范大学 Task scheduling method based on wireless metropolitan area network micro-cloud load balancing
CN110445866B (en) * 2019-08-12 2021-11-30 南京工业大学 Task migration and cooperative load balancing method in mobile edge computing environment
CN111444009B (en) * 2019-11-15 2022-10-14 北京邮电大学 Resource allocation method and device based on deep reinforcement learning

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104618480A (en) * 2015-01-29 2015-05-13 南京理工大学 Cloud system source distributing method driven on basis of network link utilization rates
CN105677447A (en) * 2016-01-29 2016-06-15 哈尔滨工业大学深圳研究生院 Clustering-based delay bandwidth minimization virtual machine deployment method in distributed cloud
US9489240B2 (en) * 2010-10-22 2016-11-08 Google Technology Holdings LLC Resource management in a multi-operating environment

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9489240B2 (en) * 2010-10-22 2016-11-08 Google Technology Holdings LLC Resource management in a multi-operating environment
CN104618480A (en) * 2015-01-29 2015-05-13 南京理工大学 Cloud system source distributing method driven on basis of network link utilization rates
CN105677447A (en) * 2016-01-29 2016-06-15 哈尔滨工业大学深圳研究生院 Clustering-based delay bandwidth minimization virtual machine deployment method in distributed cloud

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
极小通讯延迟的虚拟机分配算法;高任飞等;《计算机科学与探索》;20150828(第7期);第924-935页 *

Also Published As

Publication number Publication date
CN106874108A (en) 2017-06-20

Similar Documents

Publication Publication Date Title
Rahimi et al. MuSIC: Mobility-aware optimal service allocation in mobile cloud computing
Rahimi et al. MAPCloud: Mobile applications on an elastic and scalable 2-tier cloud architecture
CN107734558A (en) A kind of control of mobile edge calculations and resource regulating method based on multiserver
CN113011678B (en) Virtual operation platform operation control method based on edge calculation
CN108901075B (en) GS algorithm-based resource allocation method
CN102662764B (en) A kind of dynamic cloud computational resource optimizing distribution method based on SMDP
CN110069341B (en) Method for scheduling tasks with dependency relationship configured according to needs by combining functions in edge computing
CN106874108B (en) Technology for minimizing number of micro clouds in mobile cloud computing
CN111163178B (en) Game theory-based service deployment and task unloading method in edge computing
CN109947574B (en) Fog network-based vehicle big data calculation unloading method
CN110933139A (en) System and method for solving high concurrency of Web server
CN110968366B (en) Task unloading method, device and equipment based on limited MEC resources
CN110519370B (en) Edge computing resource allocation method based on facility site selection problem
Hu et al. Caching as a service: Small-cell caching mechanism design for service providers
Kapgate Weighted moving average forecast model based prediction service broker algorithm for cloud computing
CN112925607A (en) System capacity expansion and contraction method and device and electronic equipment
US20160269297A1 (en) Scaling the LTE Control Plane for Future Mobile Access
Xu et al. PDMA: Probabilistic service migration approach for delay‐aware and mobility‐aware mobile edge computing
Li et al. Load balancing edge server placement method with QoS requirements in wireless metropolitan area networks
CN107360235B (en) Task migration method based on reliability grading
Ma et al. A cyclic game for service-oriented resource allocation in edge computing
CN114077485A (en) Service scheduling deployment method for Internet of things edge computing node resources
Routaib et al. Modeling and evaluating a cloudlet-based architecture for mobile cloud computing
CN108512817B (en) Multi-video transcoding scheduling method and device
Globa et al. Architecture and operation algorithms of mobile core network with virtualization

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20210209

Termination date: 20211228